Best way of moving vPC port-channels to different interfaces

Hello,

We have a pair of 2 Nexus 3548X switches that we use together with the vPC functionality to create port-channels across the 2 switches.

We connect these switches to VM Hypervisors and each hypervisor has two port-channels: one for regular traffic (WAN/LAN) and one for migration traffic (for moving VMs between servers). We plan on moving the migration traffic to a separate switch in order to fit more servers on the same switches, as we are almost out of ports.

Moving the migration traffic isn't hard; we can just disconnect the cables and move them to the new switch.
But afterwards, of course we'll be left with a lot of empty ports on the existing switches, and we'd like to move all the remaining interfaces (used for regular traffic WAN/LAN) to the physical ports where the migration network used to be.

For example:

The current configuration is something like:

interface Ethernet1/5 switchport mode trunk spanning-tree port type edge trunk spanning-tree bpduguard enable channel-group 100 mode active no shutdown interface Ethernet1/6 switchport mode trunk switchport trunk allowed vlan 1004 spanning-tree port type edge trunk spanning-tree bpduguard enable channel-group 101 mode active no shutdown interface Ethernet1/7 switchport mode trunk spanning-tree port type edge trunk spanning-tree bpduguard enable channel-group 110 mode active no shutdown interface Ethernet1/8 switchport mode trunk switchport trunk allowed vlan 1004 spanning-tree port type edge trunk spanning-tree bpduguard enable channel-group 111 mode active no shutdown 

Here we have:

Channel group 100: WAN/LAN -> to hypervisor1
Channel group 101: Migration -> to hypervisor1

Channel group 110: WAN/LAN -> to hypervisor2
Channel group 111: Migration -> to hypervisor2

After we move the migration network to separate switches, port 6 and 8 with channel-groups 101 and 111 will be removed.

So what we'd like to do is change the configuration to:

interface Ethernet1/5 switchport mode trunk spanning-tree port type edge trunk spanning-tree bpduguard enable channel-group 100 mode active no shutdown interface Ethernet1/6 switchport mode trunk spanning-tree port type edge trunk spanning-tree bpduguard enable channel-group 110 mode active no shutdown interface Ethernet1/7 switchport mode trunk spanning-tree port type edge trunk spanning-tree bpduguard enable channel-group 120 mode active no shutdown interface Ethernet1/8 switchport mode trunk spanning-tree port type edge trunk spanning-tree bpduguard enable channel-group 130 mode active no shutdown 

So it would be:

Channel group 100: WAN/LAN -> to hypervisor1
Channel group 110: WAN/LAN -> to hypervisor2
Channel group 120: WAN/LAN -> to hypervisor3
Channel group 130: WAN/LAN -> to hypervisor4

I'm thinking of the best way to do this. I am thinking of doing it like this:

Let's say we have nexus-01 and nexus-02 for the WAN/LAN traffic

1. Remove all the physical cables we now use for the migration network from both switches (this will of course down the migration network which is ok)
2. Remove all the physical cables we now use for the WAN/LAN traffic from one of the Nexus switches; i.e. nexus-01. At this point I would expect there to be no (or a very minimal) traffic disruption as the port-channel on the secondary switch (nexus-02) would remain operational and forward traffic. I think that if we do not remove the cables at this point and started reconfiguring the interfaces there'd become a mismatch in the channel-group id and there might be a traffic disruption / port-channel going down.
3. Then from nexus-01 we remove all the port channels we used for the migration network and reconfigure the physical interfaces to have the correct channel-group (as shown in the most recent previous snippet above)
4. We then reconnect the WAN/LAN physical cables to the correct/new physical ports on nexus-01
5. Verify that all port channels are completely back up on both switches
6. Now repeat for the other switch: Disconnect all physical cables we use for WAN/LAN traffic from nexus-02 At this point there should again be no or a very minimal traffic disruption and the port-channel remains up on nexus-01
7. Repeat step 3, 4, and 5 but now for nexus-02

Does this seem like an okay way to do it? Or is there a better way / are there other things we should think of?

Thank you!